Receiver for frequency offset correction

Technical field

0001The present invention generally relates to wireless telecommunication receivers, and in particular to wireless telecommunication receivers using Bluetooth low energy (BLE).

Background of the invention

0002Bluetooth technology is a short-range communications technology for a wide range of devices ranging from mobile phones and computers to medical devices and home entertainment products. As of Bluetooth v4.0 the Bluetooth standard includes Bluetooth Low Energy (BLE) which is built upon the original key features of Bluetooth but turns to devices that operates with ultra low power wireless connectivity.

0003Any BLE receiver needs to receive dirty packets (dirty Tx) as specified in the BLE test specification document "<nplcit id="ncit0001" npl-type="s"><text>Bluetooth Low Energy RF PHY Test Specification" (RF-PHY.TS.4.1.0</text></nplcit>). Every 50 packets, the carrier frequency offset (CFO), modulation index (h), and symbol timing error (terror) are changed to specific value combinations described in the test specification and shown in table 1 below. <tables id="tabl0001" num="0001"><table frame="all"><title><i>Table 1: Transmitter parameters for dirty packets</i></title><tgroup cols="4"><colspec colnum="1" colname="col1" colwidth="18mm" /><colspec colnum="2" colname="col2" colwidth="19mm" /><colspec colnum="3" colname="col3" colwidth="18mm" /><colspec colnum="4" colname="col4" colwidth="18mm" /><thead><row><entry valign="top">Test run</entry><entry valign="top">CFO</entry><entry valign="top">h</entry><entry valign="top">t_error</entry></row></thead><tbody><row><entry>1</entry><entry>100 kHz</entry><entry>0,45</entry><entry>-50 ppm</entry></row><row><entry>2</entry><entry>19 kHz</entry><entry>0,48</entry><entry>-50 ppm</entry></row><row><entry>3</entry><entry>-3 kHz</entry><entry>0,46</entry><entry>+50 ppm</entry></row><row><entry>4</entry><entry>1 kHz</entry><entry>0,52</entry><entry>+50 ppm</entry></row><row><entry>5</entry><entry>52 kHz</entry><entry>0,53</entry><entry>+50 ppm</entry></row><row><entry>6</entry><entry>0 kHz</entry><entry>0,54</entry><entry>-50 ppm</entry></row><row><entry>7</entry><entry>-56 kHz</entry><entry>0,47</entry><entry>-50 ppm</entry></row><row><entry>8</entry><entry>97 kHz</entry><entry>0,5</entry><entry>-50 ppm</entry></row><row><entry>9</entry><entry>-25 kHz</entry><entry>0,45</entry><entry>-50 ppm</entry></row><row><entry>10</entry><entry>-100 kHz</entry><entry>0,55</entry><entry>+50 ppm</entry></row></tbody></tgroup></table></tables>

0004In addition to fixed frequency offset, a defined frequency drift is added to the signal characteristics. This is implemented by adding a sinusoidal low frequency modulation to the signal, with deviation of 50 kHz and a modulation frequency of 625 Hz. The modulating signal is synchronized with the packets so that each alternating packet starts at 0° and 180° of the modulating signal.

0005That is, a dirty Tx has both an initial carrier frequency offset (CFO) and a carrier frequency drift during transmission.

0006One of the difficulties is to timely estimate and correct the CFO as the preamble is short (1 octet) while the start of frame delimiter (SFD) needs to be received bit correct. Moreover the drift needs to be tracked during payload reception.

Summary of the invention

0007In view of the above, an objective of the invention is to provide receiver with increased capabilities to correctly estimate a CFO and a carrier frequency drift during packet transmission. This and other objects are achieved by providing a receiver and a method having the features defined in the independent claims. Preferred embodiments are defined in the dependent claims.

0008According to a first aspect of the present invention a receiver is provided. The receiver comprises a demodulator adapted to demodulate a radio frequency (RF) signal received at an input thereof and to output a demodulated signal. The receiver comprises a plurality of high-pass filters connected to an output of the demodulator for receiving and high-pass filtering the demodulated signal, wherein the plurality of high-pass filters are adapted to operate with a first set of filter responses during a first time period of the demodulated signal and operate with a second set of filter responses during a second time period of the demodulated signal.

0009According to a second aspect of the invention a method is provided for processing a received signal comprising a modulated carrier signal of a predetermined frequency, wherein the carrier signal comprises a carrier frequency offset. The method comprises demodulating the received carrier signal, receiving the demodulated signal in a plurality of high-pass filters, and high-pass filtering the demodulated signal using a first set of filter responses for the plurality of filters during a first time period of the demodulated signal and high-pass filtering the demodulated signal using a second set of filter responses for the plurality of filters during a second time period of the demodulated signal.

0010Thus, the present invention is based on the idea of providing a device and method using a plurality of filters which may be dynamically optimized for filtering different parts of the received signal. As the filters are adapted to operate with a first set of filter responses during a first time period of the demodulated signal, and operate with a second set of filter responses during a second time period of the demodulated signal, it is, at different times during reception of the signal, possible to monitor the filtered signals at the outputs of the filters and select one of the filtered signals, which have been filtered at a desired level according to one filter response of the first or second set of filter responses, for further processing.

0011By the term "receiver" is here in general meant any type of device that is capable of receiving a radio frequency signal, and in particular a device that is adapted to receive radio frequency signals that are transmitted according to the Bluetooth standard operating in the 2.4 GHz Industrial Scientific Medicine (ISM) band.

0012By the term "demodulator/demodulating" is here generally meant a unit or function in which an information-bearing signal is extracted from a modulated carrier wave. The demodulator may be implemented by electronic circuitry, computer program code portions which when executed in e.g. digital signal processor perform the function of the demodulator, or a mix thereof, which is used to recover the information content from the modulated carrier wave. In a preferred embodiment the carrier is modulated using Gaussian Frequency Shift Keying (GFSK). Other types modulation, such as Frequency Shift Keying (FSK) and variants thereof (e.g. CPFSK, MSK and GMSK), Phase shift Keying (PSK) and variants thereof (e.g. BPSK, pi/2 DBPSK, pi/4 DQPSK, OQPSK, SOQPSK and DPQPSK) may also be used. The signal output from the demodulator may be an analog or digital signal.

0013By the term "high-pass filtering means" is here meant a unit that comprises filters which passes signals with a frequency higher than certain cutoff frequencies and attenuates signals with frequencies lower than the cutoff frequencies. The high-pass filtering means may be implemented by electronic circuitry, computer program code portions, or a mix thereof. The amount of attenuation for each frequency depends on the filter design. The filter may be of a first or higher order. In a preferred embodiment each filter in the high-pass filtering means is implemented as a digital filter comprising a differentiator and an integrator as disclosed in more detail below.

0014By the term "a plurality of filters adapted to operate with a first set of filter responses" is here meant that each filter of the plurality of filters is defined by one or more filter parameters controlling the cutoff frequency and/or stop band attenuation of the filter. Thus, each filter of the plurality of filters may provide a specific response to a signal applied at its input.

0015By the term "first/second time period of the demodulated signal" is here meant a first and second period in time, each defined by a start time and a stop time, during which a first and second portion of the demodulated signal is received by the high-pass filtering means. As disclosed in more detail below, the first and second time period may be predefined time periods or may be time periods resulting from events such as the detection of a specific pattern in the demodulated signal.

0016According to an embodiment of the present invention, each filter response in the first set of filter responses has a corresponding filter response in the second set of filter responses, wherein each of the corresponding filter responses in the second set of filter responses is slower than the corresponding first filter response.

0017By providing a filter response in the second set of filter responses that is slower than a corresponding filter response in the first set of filter responses it is possible to control the amount of filtering applied to different parts of the demodulated signal. By way of example, if the content of demodulated signal during the second time period is more sensitive to filtering, i.e. is more prone to be corrupted should it be filtered too aggressively, it is hence possible to provide a first, more aggressive level of filtering during the first time period, and a second, less aggressive level of filtering during the second time period.

0018By the term "corresponding filter response" is here meant that in the first set of filter responses are a number of responses defined by filter parameters α[1], β[1], γ[1], etc., and in the second set of filter responses are a number responses defined by filter parameters α[2], β[2], γ[2], etc., wherein the filter response defined by filter parameter α[1] (β[1], γ[1], etc.) in the first set of filter responses has a corresponding filter response defined by filter parameter α[2] (β[2], γ[2], etc.) in the second set of filter responses.

0019By the term "slower" is here meant that the filter response is less aggressive such that a lower level of filtering is achieved.

0020According to an embodiment of the invention the receiver is adapted to receive an RF signal in the form of a modulated carrier signal with a predetermined carrier frequency, and wherein said demodulator is adapted to demodulate a carrier frequency offset, CFO, present in the RF signal into a DC offset.

0021An advantage of this embodiment is that subsequent processing of the received data in order correct the received data with a CFO estimate is facilitated. That is, by demodulating the CFO into a DC offset, an appropriate high-pass filter may be selected for subsequent filtering such that the CFO may be compensated for.

0022The DC offset may be represented by an analog or digital signal. The predetermined carrier frequency depends on the communication standard employed. According to a preferred embodiment the predetermined carrier frequency corresponds to a center frequency defined according to the Bluetooth low energy standard operating in the ISM frequency band located at 2400 MHz to 2483.5 MHz. In particular, the Bluetooth low energy system uses center frequencies 2402 + n * 2 MHz, where n = 0,1,2...39. Any offset from the center frequency will hence result in a DC offset corresponding to the frequency offset.

0023According to an embodiment of the invention the receiver comprises a selector adapted to select one of the plurality of filters based on the CFO.

0024An advantage of this embodiment is that based on the amount of CFO / DC offset present in the received signal an appropriate filter may easily be selected for processing the demodulated signal.

0025The selector may be in the form of a unit implemented in hardware, software or a combination of both, whose function in general terms is to provide the output signal from one of the filters to other units for subsequent processing. According to a preferred embodiment the selector comprises a measuring unit for determining the amount of CFO / DC offset present in the demodulated signal, and a multiplexing unit for selecting one of a plurality of analog or digital input signals received from the plurality of filters and forward it to its output. The multiplexing unit may comprise one or more controllable switches, e.g. analog switches or digital gates, which are controlled by signals received from the measuring unit.

0026According to an embodiment of the invention the demodulator is a differential phase-domain demodulator.

0027By providing a differential phase-domain demodulator a CFO present in the received signal will be directly demodulated into a DC offset at the output of the demodulator. Thus, the complexity of the receiver will be reduced since no further processing of the demodulated signal, in order to derive a DC offset signal, is needed before providing the demodulated signal to the high-pass filtering means.

0028According to an embodiment of the invention the demodulated signal is in the form of a data packet.

0029By providing a receiver comprising a demodulator adapted to demodulate the received signal into data packets, the selection of appropriate filters and filter responses for processing different parts of the demodulated signal is facilitated. That is, a first filter operating with a first filter response may be used for filtering a first part of the packet depending on the assumed content of the first part of the packet (such as a preamble), and a corresponding (or different) filter operating with a second filter response may be used for filtering a second part of the packet (such as a synchronization word or payload of the packet).

0030The data packet may take various forms depending on the communication standard employed. By way of example, a standard BLE packet comprises a preamble (1 octet), an address word (4 octets), PDU (2-39 octets) and CRC (3 octets). For testing purposes a packet format as defined in the test specification document is used, wherein the reference packet comprises a preamble (1 octet), a synchronization word (4 octets), a PDU header (1 octet), PDU length data (1 octet), PDU payload (37 octets) and CRC (3 octets). Other packet formats, such as employed in IEEE802.15.4 (ZigBee) or IEEE802.15.6 (Wireless Body Area Network) may also be used.

0031According to an embodiment of the invention the high-pass filters may be adapted to operate with the first set of filter responses on a first part of the data packet and the second set of filter responses on a second part of the data packet.

0032An advantage of this embodiment is that different parts of the packet may be more or less aggressively filtered. Thus, payload data in the packet, which likely is more sensitive to filtering, i.e. is more prone to be corrupted should it be filtered too aggressively compared to e.g. the preamble, may be subject to a less aggressive filter operating with a slower set of filter responses.

0033By the terms "first part" and "second part" of the packet are here meant either parts of the packet strictly related to certain time periods from e.g. the start of the reception of the packet, i.e. the first part may correspond to the part of the packet received during the first 8 µs of reception, while the second part may correspond to the upcoming 32 µs, or parts of the packet related to the content of the packet, e.g. the first part may relate to the preamble of the packet and the second part may relate to the synchronization word in the packet, e.g. a first part may be a part of the packet starting at the detection of the start of the preamble (also referred to as "energy detect") and ending at the detection of the synchronization word, and the second part may be a part of the packet starting at the detection of the synchronization word and ending at the detection of the PDU header. A combination of both previous examples is also within scope of the invention, e.g. a first part may be a part of the packet starting at the detection of the start of the packet and ending after a certain amount of time, and the second part may be a part of the packet starting at the detection of the synchronization word and ending at the detection of the PDU header.

0034According to an embodiment of the invention the receiver may comprise a buffer connected to the output of the demodulator for buffering at least a portion of the demodulated signal.

0035An advantage with this embodiment is that a part of the demodulated signal may be acted upon while in the buffer. That is, by buffering a portion of the demodulated signal, a processing unit may alter the content of the buffer before outputting the content of the buffer to subsequent processing.

0036By the term "buffer" is here meant a unit that is adapted to temporarily store data in analog or digital form. According to a preferred embodiment, the buffer is a shift register adapted to store a digital word corresponding to the length of the synchronization word in the test signal.

0037The buffer may be implemented by electronic circuitry (a hardware shift register), computer program code portions which when executed in e.g. digital signal processor perform the function of the buffer (storing the data in a memory), or a mix thereof.

0038The receiver may comprise a comparator connected to the output of the demodulator and an output of the high-pass filters, and adapted to provide a correction signal to an input of the buffer based on a difference between the demodulated signal and a filtered demodulated signal.

0039An advantage with this embodiment is that the content of the buffer may easily be corrected based on the amount of CFO present in the demodulated signal. This is especially beneficial since it becomes possible to correct the first samples of the demodulated signal, fed to the buffer, with the latest CFO estimate.

0040The "comparator" may be in the form of a unit implemented in hardware, software or a combination of both, whose function in general terms is to provide a signal indicative of the difference between the demodulated signal and a filtered version of the demodulated signal. That is, at any time instant the content of the buffer may be corrected with the most recent CFO estimate.

0041The inventive method may comprise: receiving an RF signal in the form of a modulated carrier signal with a predetermined carrier frequency, and demodulating a carrier frequency offset, CFO, present in the RF signal into a DC offset.

0042An advantage of this embodiment is that subsequent processing of the received data in order correct the received data with a CFO estimate is facilitated. That is, by demodulating the CFO into a DC offset, an appropriate high-pass filter may be selected for subsequent filtering such that the CFO may be compensated for.

0043The inventive method may comprise selecting one of the plurality of filters based on the CFO.

0044An advantage of this embodiment is that based on the amount of CFO present in the received signal an appropriate filter may easily be selected for processing the demodulated signal.

0045The inventive method may comprise buffering at least a portion of the demodulated signal and correcting the buffered at least a portion based on a difference between the demodulated signal and the high-pass filtered demodulated signal.

0046An advantage with this embodiment is that a part of the demodulated signal may be acted upon while in the buffer. That is, by buffering a portion of the demodulated signal, the buffered signal may easily be corrected based on the amount of CFO present in the demodulated signal. This is especially beneficial since it becomes possible to correct the first samples of the demodulated signal, fed to the buffer, with the latest CFO estimate.

0047According to an embodiment of the present invention a receiver is provided comprising: a demodulator adapted to demodulate a radio frequency, RF, signal received at an input thereof and to output a demodulated signal, high-pass filtering means connected to an output of the demodulator for receiving the demodulated signal, wherein the high-pass filtering means is adapted to operate with a first filter response during a first predetermined time period of the demodulated signal and operate with a second filter response during a second predetermined time period of the demodulated signal

0048Other objectives, features and advantages of the present invention will appear from the following detailed disclosure, from the attached claims as well as from the drawings.

0049Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a/an/the [element, device, component, means, step, etc]" are to be interpreted openly as referring to at least one instance of said element, device, component, means, step, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.

Brief description of the drawings

0050The above, as well as additional objects, features and advantages of the present invention, will be better understood through the following illustrative and non-limiting detailed description of preferred embodiments of the present invention, with reference to the appended drawings, where the same reference numerals will be used for similar elements, wherein: <ul id="ul0001" list-style="none" compact="compact"><li><figref idref="f0001">Fig 1</figref> is a schematic block diagram of a receiver according to an embodiment of the present invention;</li><li><figref idref="f0001">Fig 2</figref> diagrammatically illustrates a test packet according to the BLE test specification document;</li><li><figref idref="f0002">Fig. 3</figref> diagrammatically illustrates the use of different filter parameters;</li><li><figref idref="f0002">Fig. 4</figref> is a schematic block diagram of a filter selection unit according to an embodiment of the present invention;</li><li><figref idref="f0003">Fig. 5</figref> diagrammatically illustrates the convergence speed of different filters according to an embodiment of the present invention;</li><li><figref idref="f0004">Fig. 6</figref> is a schematic block diagram of a high-pass filter according to an embodiment of the present invention.</li><li><figref idref="f0005">Fig. 7</figref> is a schematic block diagram of a buffer arrangement connected to the output of the demodulator for buffering at least a portion of the demodulated signal.</li><li><figref idref="f0006">Fig. 8</figref> illustrates measurement results obtained by a receiver according to an embodiment of the present invention</li><li><figref idref="f0007">Fig. 9</figref> illustrates simulation and measurement results obtained by embodiments of the present invention.</li></ul>

Detailed description of preferred embodiments

0051<figref idref="f0001">Fig 1</figref> is a block diagram illustrating an overview a receiver 100 according to an embodiment of the present invention. The in-phase (I) and quadrature (Q) signal components are preferably converted to digital format in a pair of analog-to-digital converters (ADC) 101a, 101 b. The digital signals are then converted into the phase domain by PM block 102 and further into differential phase domain by differential block 103 for further processing of the signals in the differential phase domain, which is equivalent to the FM domain. A benefit of processing in the differential phase domain is that the carrier frequency offset (CFO) becomes a direct current (DC) offset in the differential phase domain.

0052A CFO correction block 104, which is disclosed in detail below, receives the signals from the differential block 103 and its output signal is provided to a bit decision block 105.

0053<figref idref="f0001">Figure 2</figref> illustrates a typical test packet 200 used according to the test specification document. As indicated above, the packet comprises a preamble (1 octet) 201, a synchronization word (4 octets) 202, a PDU header (1 octet) 203, PDU length data (1 octet) 204, PDU payload (37 octets) 205 and CRC (3 octets) 206.

0054As the CFO is changed during a test of the receiver 100, a challenge for any receiver is to estimate and compensate for the CFO. This is especially important for BLE applications since the preamble 201 is very short, i.e. 8 symbols which correspond to 8 µs at a transfer rate of 1 Mbps. Additionally, the carrier can drift during reception of the PDU payload 205.

0055With reference to <figref idref="f0002">figures 3 and 4</figref>, to compensate for the CFO the present invention employ a DC offset cancellation strategy in the differential phase domain. As mentioned above, the CFO manifests itself as DC offset in the differential phase domain. This makes it possible to filter the demodulated signal via a set of high-pass filters 404a-c. As disclosed in detail below, to increase the convergence speed, the filters 404a-c are gear shifted depending on which part of the packet that is currently processed.

0056When the receiver is enabled, initial filter parameters are loaded 301 into the filters 404a-c. Typically the filters are set to "transparent", indicated by the filter parameters α[0], β[0], γ[0] in <figref idref="f0002">Fig. 3</figref>, which means that by loading parameters α[0], β[0], γ[0] into the filters 404a-c, the response of the filters 404a-c will be such that a signal received at the input of the filters will pass through the filters unaffected (i.e. unfiltered).

0057At energy detect, i.e. at detection of the preamble 201, the filters 404a-c are gear shifted, i.e. new filter parameters α[1], β[1], γ[1] are loaded 302 into the filters 404a-c. The filter parameter α[1] defining filter "HPF α" 404a in <figref idref="f0002">Fig. 4</figref> is used for moderate CFO, filter parameter β[1] defining filter "HPF β" 404b is used for intermediate CFO, and filter parameter γ[1] defining filter "HPF γ" 404c is used for large CFO. As illustrated in <figref idref="f0003">Fig. 5</figref>, the filter parameter γ[1] provides a faster convergence of filter "HPF γ" 404c compared to filter parameter β[1] of filter "HPF β" 404b, which in turn has faster convergence than filter "HPF α" 404a defined by filter parameter α[1].

0058The outputs from the filters 404a-c are monitored by a filter selection unit 404d which determines the slowest of the filters 404a-c that is able to detect the CFO and provides the output of that filter for further processing via a multiplexer 404e. Thus, by the provision of the different filters 404a-c, and the selection of the slowest possible filter, it is possible to compensate for the CFO during the processing of the first part of the packet while at the same time minimizing the impact of the filtering on the demodulated signal.

0059A period after energy detect, at the start of the synchronization word 202, new filter parameters α[2], β[2], γ[2] are loaded 303 into the filters 404a-c (i.e. the filters are gear shifted). Each of the new filter parameters provides a filter response that is slower compared to what is provided with the corresponding filter parameter loaded into the filters during reception of the preamble 201. That is α[2] provides a slower filter response than α[1], β[2] a slower filter response than β[1], and γ[2] a slower filter response than γ[1]. By this provision the damaging effect of the filtering of the demodulated signal is mitigated. That is, due to its content, the preamble 201 is less prone to being corrupted compared to the synchronization word 202, wherein a more aggressive filtering of the preamble in order to detect the CFO may be employed.

0060When the synchronization word 202 has been received, i.e. at the start of the PDU 203, new filter parameters α[3], β[3], γ[3] are loaded 304 into the filters 404a-c. Each of the new filter parameters preferably provides a slower filter response than the corresponding filter parameter loaded into the filters during reception of the synchronization word 201. In an alternative embodiment each of the new filter parameters provides a faster response than the corresponding filter parameter loaded into the filters during reception of the synchronization word 201. In yet an alternative embodiment the filter parameters are not altered after reception of the synchronization word 201.

0061In the description above, an embodiment employing three different filters has been disclosed. It is, however, emphasized that more or less than three filters may be used, e.g. two, four, five, six etc. depending on how many different levels of convergence speed that is desired. By way of example, if the present invention is employed for reception of a packet sent via the ZigBee protocol, only two filters will be needed due to the structure of the ZigBee package.

0062<figref idref="f0004">Fig. 6a</figref> is a schematic block diagram of a high-pass filter 600 according to an embodiment of the present invention. In general terms, the filter 600 comprises a differentiation block 601 connected to an integration block 602. The frequency response of the differentiation block 601 is illustrated by the dashed line 610 in <figref idref="f0004">Fig 6b</figref> and the frequency response of the integration block 602 is illustrated by the dashed line 620 in <figref idref="f0004">Fig. 6b</figref>. The cascade coupling of the two blocks provides a total frequency response of the filter 600 as illustrated by the solid line 630 in <figref idref="f0004">Fig. 6b</figref>.

0063The feedback loop of the integrator 602 comprises a multiplication block 603 into which the filter parameters α, β, γ are loaded. Thus by providing different filter parameters, the total frequency response of the filter 600 may be altered such that the cut-off frequency of the filter is lowered or raised. Thus, by the provision of a filter structure as shown in <figref idref="f0004">Fig. 6a</figref>, fast and simple gear switching of the filter 600 is provided.

0064<figref idref="f0005">Fig. 7</figref> is a schematic block diagram of a buffer arrangement connected to the output of the demodulator for buffering at least a portion of the demodulated signal. In a preferred embodiment the buffer 701 is formed by a shift register, the input of which is connected to the output of the demodulator for receiving the demodulated samples. As the demodulated samples are fed into the shift register 701, the data in the shift register 701 is corrected with the most recent CFO estimate received from the CFO correction block 104. That is, by comparing the filtered and unfiltered data, an estimation of the CFO present in the signal may be estimated and continuously subtracted from the data in the shift register 701 during reception of the packet. Thus, the first samples in the shift register may be corrected with the latest known CFO estimate, effectively increasing the convergence time for the SFD. The correction of the data in the shift register is continuous in the sense that it may be performed once the shift register has been filled with data, or performed on blocks of data in the shift register as samples are fed into the shift register.

0065Finally, hard detect of the symbols is done based on the sign information of the corrected data in the shift register. A received synchronization word is correlated against the expected synchronization word / device address in order to determine if the synchronization word is correctly received and to determine if the packet is destined to the present receiver. The receiver returns to detecting a new preamble should the correlation show that the words do not match, otherwise the receiver continues to receive and CFO-correct the payload data of the packet.

0066Packet Error Rate (PER) measurement is used in all measurements testing receiver characteristics in the test specification document.

0067The packet error rate (PER) is defined as: <maths id="math0001" num=""><math display="block"><mrow><mi mathvariant="italic">PER</mi><mo>=</mo><mfenced separators=""><mn>1</mn><mo>−</mo><mfrac><mi mathvariant="italic">Number of correctly received packets</mi><mi mathvariant="italic">Total number of transmitted packets</mi></mfrac></mfenced><mo>*</mo><mn>100</mn></mrow></math><img file="EP3244584A1_D0001.tif" /></maths>

0068The sensitivity level based on bit error rate (BER) measurements is defined as the input power level at which a BER of 0.1% is achieved measured with a reference signal as described in the test specification document and packet with PRBS9 payload as described in "<nplcit id="ncit0002" npl-type="s"><text>Bluetooth Specification v4.0, Vol 6, Part F; Direct Test Mode</text></nplcit>".

0069This PER requirement equates to a BER of 0.1 % under the assumption that bit errors are randomly distributed with a rectangular error probability density function, and that bit errors are not correlated. Furthermore, the probability of a particular bit being in error at a BER of 0.1 % is 0.001, from which it follows that the probability of a bit being OK under the same condition is 0.999.

0070Examining the impact of a bit error in the test packet gives that a packet is lost if a bit error is present in any part of the packet, save the 8-bit preamble. That is, the number of significant bits in an test packet is thus 368 bits (out of a total of 376 bits). Further, the probability of a 368 bit sequence containing no bit errors is 0,999³⁶⁸ = 0,692, and hence the resulting PER requirement becomes (1 - 0.692) * 100% = 30.8%

0071<figref idref="f0006">Fig. 8</figref> illustrates measurement results 801 obtained by a receiver according to an embodiment of the present invention. As can be seen from the figure, the PER is well below the required 30.8% level 802 for a CFO measurement of +/-180 ppm, while the requirement for passing the test is that a range of +/-62 ppm needs to be covered.

0072<figref idref="f0007">Fig. 9</figref> illustrates simulation 901, 902 and measurement 903, 904 results obtained by the present invention. As can be seen from the figure, the present invention provides a PER well below the required 30.8% 905 level for signal levels falling below -90 dBm.

0073Embodiments of the present invention may be implemented in, for example, integrated circuits or chip sets, wireless systems, and receiver system products. For example, a processor is operative to execute software adapted to perform the demodulation techniques of the present invention. Demodulation software is adapted to reside on a computer readable medium, such as a magnetic disk within a disk drive unit. The computer readable medium may also include a flash memory card, EEROM based memory, ROM storage, etc. The software adapted to perform the demodulation method may also reside, in whole or in part, in the static or dynamic main memories or in firmware within a processor (i.e. within microcontroller, microprocessor or microcomputer internal memory). The demodulation method may also be applicable to implementations in integrated circuits, field programmable gate arrays (FPGAs), chip sets or application specific integrated circuits (ASICs), wireless systems, and other communication system products.

0074The invention has mainly been described above with reference to a few embodiments. However, as is readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the invention, as defined by the appended patent claims.