design a simplified kalman filter in a vhdl module

Data is not only subject to noise affecting the transmission process but also originates in the instrument creating the data. Most noise appears stochastic –i.e., random and asynchronous to the data sampling rate. Thus, noise can be removed by filtering out portions of the signal at frequencies higher
than the known frequency band of the data source. However, long term averaging severely reduces the response speed. One technique, known as “Kalman filtering”, also known as linear quadratic estimation (LQE), is an algorithm that uses a time series of measurements to produce estimates of unknown data that tend to be more accurate than those based on a single measurement alone. In its simpler form, a Kalman filtered signal is the weighted average of a certain number of previous samples and the current sample. To provide quick response time, the filtered signal is updated as quickly as possible with each new datum. Kalman filtering is used extensively in navigation. It was used for the nonlinear problem of trajectory estimation for the Apollo program leading to use in the Apollo navigation computer enabling the Moon landings. Kalman filters support the navigation systems of U.S. Navy nuclear ballistic missile submarines and the guidance and navigation systems of cruise missiles. It is also used in the guidance and navigation systems of reusable launch vehicles and the attitude and navigation systems of spacecraft docking at the International Space station.

Your mission: Design a simplified Kalman filter in a VHDL module.
A data stream of bytes is being received synchronous to the local system clock: 10 megabytes/sec, as unsigned bytes. With each new datum, the filter must:

1) Compute the weighted average of the previous 3 samples and the current sample before the next clock edge. The result is a Moore output with no glitches.

2) The four samples are weighted 4, 2, 1, 1. I.e. let s(3) be the current sample, s(0) the oldest, Output = ∑ ( 4 * sample (3) + 2 * sample (2) + sample (1) + sample (0) ) div 8.

3) Multiplication and division is intentionally a power of 2 to enable calculation by shifting bits.

4) When the next datum arrives, sample (0) <= sample (1), sample (1) <= sample (2), sample (2) <= sample (3), sample (3) <= new_datum and the process repeats.

 Create the VHDL module by first creating a state chart with RTL notation.

 Speed of calculation is the prime mission; register / logic minimization is not the mission.

 Accomplish the multiplication and division by bit shifting – e.g. using the sll operation.

Notes:

 Use type unsigned for the sample values. For convenience, create an “array of unsigned”. Specify a range that covers at least byte values * 8 to preclude overflow / carry events.

 At each clock edge, the first process step is to capture the previous calculated output value into an output register, so the clock can be as fast as possible yet provide a Moore output.

 The calculation should be accomplished in combinatorial logic. Ergo the next process step is to shift register contents (sample (0) <= sample (1), etc.) and capture the new datum. The result of the combinatorial calculation is awaiting capture in a register at the next clock.

Estimate the minimum clock period for this filter: shifts take 10 ns, addition takes 25 ns. Be sure to consider what operations occur simultaneously. A timing diagram will be useful.