1. Title of the invention

[001] Robust Squelching Technique for Airborne AM/FM receiver through carrier detection at low CNR condition.

2. Field of invention

[002] The invention is related to the field of Aeronautical communication in legacy AM & FM Modes and signal processing.

3. Background of invention

[003] The purpose of SQUELCH is to detect the presence of VOICE, TONE and NOISE in airborne radio receiver and muting the radio receiver Audio output in case NOISE is detected and un-mute the Audio in case Speech and tone is detected. The process of audio muting through Squelching, is. presented here. (3,. 4,. 5 Figure-1). The. reliable operation of Squelch is utmost important for pilots while flying an aircraft or helicopter as any unwanted receiver audio (Noise) output disturbs the pilot and at the same time none of the voice calls from ATC or any other aircrafts should be missed. The specification of the Squelch feature in the airborne radio receiver are as given below.

• Distinguish a voiced transmission as well as tones in an RF (Radio Frequency), interfering environment mixed, with. RF Noise.

• Detection decision should be fast enough (Squelch Cut-in, detection time < 40 msec) so that initial portion of speech is not missed.

• Once Detected the decision should hold for a Practical speech pause of minimum 500 msec.

• Audio should be muted fast enough (Squelch Cut-out, less than 40 msec) after a Voiced call is completed i.e. transmission is stopped.

• Detection should be reliable for speech signal mixed with white or colored noise as well as speech mixed with tones.
[004] Further in the prior art (fig-4), legacy airborne transceivers works based on absolute Signal to Noise Ratio (SNR). Here, energy of Audio frequency of 0.3-3.3 kHz is taken as Signal band energy and Audio frequency of 3.3 kHz-6.6 kHz is taken as Noise band energy. During the presence of voiced speech, low frequency audio energy is present in Signal, band. in. more, quantity than compared to. Noise, band.. The Ratio, of Signal band energy to Noise band energy is calculated and termed as SNR (Signal to Noise Ratio). A minimum of 6dB SNR is decided as threshold for the voiced speech. In present day airborne platforms EMI (Electro Magnetic Interference) from other instruments in aircraft and leakage from various radio communications have significantly increased. Conventional SNR (Signal to Noise Ratio) calculation to detect the presence, of voice for squelching, has. proven, ineffective. A wrong Squelch output such as a false alarm or missed detection in a noisy environment in V/UHF airborne radio will baffle pilots during the critical flight scenario. V/UHF airborne radios are expected to operate in a wide dynamic range of Radio frequency signal strength (- 0 dBm to -101 dBm) over a wide range of frequencies. During the absence of a valid intended signal the legacy FM (Frequency Modulation) /AM (Amplitude Modulation) channel output generates aloud-vexing noise. Therefore,, a robust and reliable Squelch is. required to distinguish the presence of valid voice/ tone from Noise at Low CNR (Carrier to Noise Ratio) as well as SNR (Signal to Noise Ratio) condition.

[005] PATENT JP2012019444A “Receiver”

The patent here describes, about. Squelch process, comprising of carrier detection and Noise power detection with dynamic threshold. However, the carrier detection scheme will not be effective in case of offset in Receiver frequency. Carrier detection method requires carrier to be tracked for proper estimation. There is also no mention of data smoothing or averaging for avoiding fluctuations at Low CNR (CARRIER TO NOISE
RATIO) condition. These deficiencies in this literature makes it difficult to use this scheme in airborne application.

[006] PATENT CN103975607

“Automatic squelch method for wireless receiving voice”

The patent here discussed about a similar squelch process which comprises of Carrier detection through carrier tracking and searching, however does not mention anything about smoothing process of the carrier signal data for avoiding fluctuation at Low CNR (Carrier To Noise Ratio).The patent here also describes about finding Modulation index as a measure of presence of Audio, which will not work in presence of Narrow band RF (Radio Frequency) interferer. The patent here also has no mention about handling of speech pauses during squelching. The cited patent here also have no reference for Squelching regarding FM (Frequency Modulation) signal and moreover the suggested scheme for estimation of estimation modulation degree for SNR (Signal to Noise Ratio) estimation will not work for in case of FM.

4. Brief Summary of Invention

[007] It is accordingly, an object of present invention is to develop a Robust Squelching Technique for Airborne AM /FM receiver with high probability of detection and very low probability of False alarms that comprises of a Novel carrier detection scheme (3,Fig-1) to estimate carrier energy by means of carrier tracking of AM /FM signal, a narrowband low pass filter (9,Fig-2) followed by a moving average filter (10,Fig-2) and also to noise energy estimation by calculating energy of the entire 20 KHz band (7, Fig-2) through the similar moving average filter (11, Fig-2), whenever the CNR is below threshold. The CNR (Carrier To Noise Ratio) value is calculated based on the ratio of Carrier energy and Noise energy (20, Fig-2) and carrier present flag is set based on the comparison of CNR value to. the set threshold. (.1.3,14, Fig-2)..
[008] The said Carrier detection scheme (3, Fig-1) is further coupled with a SNR (Signal to Noise Ratio) estimation scheme (4,Fig-1) in the baseband audio stage (15, Fig-3) which estimates the signal energy in the lower band (16,18, Fig-3) band and noise energy in the higher band (17,19,Fig-3). The results of SNR estimation and carrier present flag are coupled (22, Figr3.).to. arrive.at a. Squelch decision (5,Fig-1).

5. Brief description of drawings

[009] The foregoing and other features, objects and advancement of the present invention are apparent over the prior art in the following more particular description of preferred embodiment of the invention as illustrated in the accompanying block diagrams in the form of drawings wherein each embodiment is represented by a numeral'.

[010] Fig: 1 shows the block diagram of present scheme proposed. The Digital IF (Intermediate Frequency) samples are collected (6) and passed to channel filter to band limit the IF (Intermediate Frequency) signal to 20 KHz (1).The filtered signal is passed to both demodulator (2) as well as carrier detection scheme (3). The demodulated audio baseband is fed. to. Squelch decision, process where SNR (Signal, to. Noise Ratio) estimation and Carrier present flag is coupled and a final squelch decision is arrived at (4).The Squelch decision flag is sent to audio muting circuit for muting the output or allowing the output based on Squelch decision (5).

[011] Fig: 2 This figure depicts the block diagram of Carrier detection process. Here two paths are shown for estimating the carrier energy (10) and. Noise, energy (.1.1.). The costas loop tracks, the carrier and brings the signal to Zero IF (8).The narrow band LPF (Low Pass Filter) placed after carrier tracking provides Carrier samples (9). The fast attack slow decay Moving average filters (10,11) smoothes the data in both carrier and Noise path respectively for avoiding fluctuations at Low CNR (Carrier To Noise Ratio). The estimation of CNR (12) is done by taking ratio of Carrier and
Noise energy. The CNR is compared with fixed threshold (13) to set the carrier present flag (23). The CNR is also compared (14) to threshold with hysteresis for resetting the carrier present flag (23).

[012] Fig:3 This figure depicts the process of SNR estimation and final squelch decision making process (22). Here Baseband Audio after demodulator (15) is fed to Signal energy path and Noise Energy path in parallel. Signal energy is calculated after the filtering the audio by a BPF (BAND PASS FILTER) of 0.3 to 3.3 KHz (16) and then passing the signal by. a Fast attack slow decay Moving, Average, filter (1.8).Noise, energy is. calculated after the filtering the audio by a HPF of 6.25 KHz (17) and then passing the signal by a Fast attack slow decay Moving Average filter (19).SNR (Signal to Noise Ratio) is estimated (22) and then compared with threshold (21) to set the SNR ok flag (26).The SNR is also compared to threshold with hysteresis (25) to reset the SNR ok flag (26). Finally Squelch decision (27) is arrived at by coupling (22) SNR ok flag (26) and carrier detection.flag (23,28)..

[013] Fig:4 This figure depicts the prior art of Squelch process which only comprises of SNR (Signal to Noise Ratio) estimation, generation of final squelch decision(24) and muting the final audio (29). Here No additional information from RF (Radio Frequency) or IF (Intermediate Frequency) Demodulator (23) stage is provided to Squelch decision for robust decisions.

6. Detailed Description of Invention

[014] A novel squelch method for Legacy AM (AMPLITUDE MODULATION)/FM (Frequency Modulation) mode in an airborne V/UHF receiver is proposed. The squelch process comprises of Novel carrier detection scheme in Digital IF (Intermediate Frequency) stage coupled with, a SNR (Signal to Noise Ratio), estimation scheme, in. baseband, audio, stage. (Fig-1)
[015] The Digital IF (Intermediate Frequency) samples are

collected through a High Speed ADC(Analog To Digital Converter) at (6) and then the samples are down-converted and fed to carrier detection process (3). The carrier detection process first tracks the carrier using the Costas loop based PLL (Phase locked loop) (8). The signal for FM is given by. equation, t

j(t) j4ccos{(u)c -t- A)t + k I m(t)dt)

h

The same for AM (AMPLITUDE MODULATION) is given by equation.

5(0 = AC{1 +- cos(a>c + A)t

The costas loop tracks the carrier by correcting frequency error A in the given equations. The tracked carrier is then fed to a narrow band Low pass filter with cut off 30 Hertz for AM and 150 Hertz in case of FM (9). The filtered output is. then. fed. to. a. fast attack,, slow decay moving average filter for smoothing the data to avoid fluctuation. The equation used for the same is y (n) = coef * v(n) + (1-coef) * y(n-1), Where v (n) is absolute instantaneous signal value, y(n-1) and y(n) are previous and present value of smoothed data, 'coef is the coefficient to control attack and decay rate, 'coef value in the range of 0.8 to 0.95 is used. The smoothed signal y(n) is averaged over 2000 samples i.e. 20 msec at 100 KHz sampling rate and carrier energy value is obtained (10).

[016] Noise energy is estimated dynamically using aforesaid Moving average filter periodically every 20 msec during quite period i.e. CNR (Carrier To Noise Ratio) below threshold by averaging Noise power over 2000 samples at 100 KHz sampling rate over the entire channel bandwidth of 20 KHz (11). CNR is estimated dynamically every 20 msec by dividing estimated carrier power to the Noise power (12). The estimated Carrier to Noise power ratios are compared to a predefined threshold of 10 dB for AM (Amplitude Modulation) and 4 dB for FM (Frequency Modulation).The Carrier detection flag is set in case CNR is above the threshold and the flag is reset with a hysteresis of 2 dB(13,14). The carrier detection flag
status is shared with Squelch process based on CNR value comparison with the set threshold (22,23).

[017] The proposed scheme for squelch also comprises of SNR (Signal to Noise Ratio) estimation at Baseband Audio stage (15) sampled at 25 KHz after demodulation. The demodulated audio is passed through a BPF (BAND PASS FILTER) (16) of 0.3 to 3.3 KHz and the output is smoothed using previously mentioned moving average filter. The average is calculated over 800 samples at 25 KHz sampling rate i.e. every 20 msec. The output here is termed as signal energy. The demodulated audio in parallel is also fed to filter of 6.25 KHz HPF (17) for estimating the Noise energy. The Selection of 6.25 KHz as Noise band cutoff is a inventive step as in. the prior art (24,Fig-4) the Noise band is taken from HPF from 3.3 KHz. The 3.3 KHz to 6 KHz band contains high pitch noise from engines as well as harmonic distortion from Audio bands and thus band from 3.3 KHz to 6 KHz is avoided for Noise energy calculation. The output from HPF is averaged using previously mentioned moving average filter over same 800 samples and a new output is generated every 20 msec. The output here is termed as Noise energy (19). The ratio of Signal energy and Noise energy is calculated and defined as SNR (2Q),The SNR (Signal to Noise Ratio) is compared to a fixed threshold of 6 dB and SNR ok flag is set if SNR (Signal to Noise Ratio) is more than 6 dB. SNR ok flag is reset in case SNR (Signal to Noise Ratio) falls below 4 dB (21,25). The hysteresis of 2 dB is provided for protection against fluctuation along with fast attack slow decay filters. This hysteresis and smoothing through moving average filters here also aids in maintaining the squelch decision in. case. of. grammatical pauses in speech.

[018] The squelch decision is finally arrived at by logical ANDING (22) of the Carrier Detection Flag and SNR Ok flag. The squelch decision is to allow the audio i.e. Squelch cut-in in case of both the flags are set and squelch decision is negative i.e. muting of audio, Squelch cut-out in case
any one of the flag is reset. The proposed scheme meets the requirement of fast squelch cut-in and squelch cut-out because as a new decision in Carrier detection is taken every 20 msec.The Signal and Noise energy values in baseband audio stage are reset to Zero whenever the Carrier ; detection flag is reset. The squelch decision as claimed provides very high probability of. Detection (Q..98) and very low probability of missed detection as well as low probability of false alarm.