Code-directed synchronisation scheme for burst spread spectrum communications

• Main Authors: Gossink, D.E., Cook, S.C., Asenstorfex, J.A.
• Format: Conference Proceeding
• Language: English
• Subjects: Autocorrelation; Codes; Delay estimation; Maximum likelihood estimation; Modulation coding; Optimization methods; Performance gain; Signal processing; Spread spectrum communication; Timing
• DOI: 10.1109/MILCOM.1997.646690

One of the challenges in employing direct sequence spread spectrum (DS-SS) waveforms for burst-mode communication applications is the acquisition of chip/symbol synchronisation. The traditional technique of using a known preamble, which can be used to synchronise the receiver, incurs an unacceptable overhead when a low number of information bits are to be sent. Thus it becomes desirable to acquire synchronisation directly from the information bearing signal. In this paper we assume that the timing of the burst is not known but lies within an observation interval (block). The established approach for estimating symbol synchronisation for traditional DS-SS is known as "block signal processing". The performance limitations of this method are primarily a result of the underlying modulation and the fact that maximal length sequences have no coding gain. We propose a maximum likelihood synchronisation technique called the code-directed method, based on concepts derived from coding theory, that uses purpose-designed codes jointly optimised for cross and auto-correlation properties. The code directed spread spectrum receiver structure is a maximum likelihood joint estimator of both the data and the burst delay. In this way it avoids the problems that data modulation imparts on symbol timing synchronisation as with conventional DS-SS schemes. Simulations, for a specific example with a packet size n of eight bits and spreading L of 15 chips per bit are discussed.