Friday, June 18, 2010

CSA: From Ockham's (Occam's) Razor to Checking Mast

CSA: From Ockham's (Occam's) Razor to Checking Masts

Creating a 'de facto' standard is always going to be a hard job and none more so than when dealing with 'Cell Site Analysis', abbreviated to the acronym 'CSA'. The objective of the analysis is, as best possible; to determine a likely or approximate location of a particular mobile station (MS) at the material date and time of a call. There are a large and varied number of issues to be considered when entering into cell site analysis. Cell site analysis, or CSA, is not a precise science and this has largely lead many people to misconstrue how cell site analysis can be conducted and for those who require to make an interpretation about the interpretation being given by an expert, giving evidence, leads to mistrust about cell site analysis. So it all becomes rather a vicious circle of events, with few conceding their comprehension about the fragility and instability of the stance they have adopted.

CSA is a highly intelligent science, and an evolving forensics science at that, also. CSA has many elements in its foundation that are based on scientifically proven facts. For instance, the scientifically proven fact that 0-dBm (deciBel milliwatt) always equals 1-mW (milliWatt) of power (energy) is but one good example. Furthermore, such a scientific fact allows experts to make the declaration that each result in the measurements obtained are 'absolute' and can be demonstrated as 'relative' when compared alongside other 'absolute' results.

Given that CSA extends beyond obtaining measurements and extends also into the arena of the radio spectrum, radio protocols, beamforming etc and the infrastructure required to propagate and provide a service, this, too, is an area where many scientifically proven facts and mandatory requirements exists. This again leads to forming conclusions without the need of the expert to make assumptions.

The area where CSA needs assistance is to rely upon human intervention and that requires having deep knowledge of the subject matter and solid skillsets. Again, this does not require the person to make assumptions, but to demonstrate the possibilities and potential conclusions that may help inform a Court in order that the Court can arrive at its own conclusions.

There are indeed some useful scientific philosophy dictums that can aid and support a CSA practitioner that can be adopted when striving for the aim of being 'objective', in addition to ‘independent’ and ‘impartial’, and one of the most important of these is Ockham's (Occam's) Razor attributed to the distinguished 14th century medieval logician and philosopher William of Ockham (c1285-1349). 'Leff, Gordon'; in his 1958 work 'Medieval Thoughts: St Augustine to Ockham' enunicated the so-called Ockham's Razor as 'entities ought not to be multiplied except of necessity'. However, the use of the term 'Razor' in reference to a rather superficially simple phenomenon having a complex mechanism behind it, did not appear until after Ockham's death and, although he didn't invent it, it is the frequency, apparently, with which he used the phenomenon 'should make as few assumptions as possible' in his writings that associates Ockham to this dictum. This can be clearly seen from the dictum commonly associated with the Ockham (Occam's) Razor 'Numquam ponenda est pluralitas sine necessitate', translated, means, 'Plurality ought never be posited without necessity'. How incredible, within six words, he encapsulates in that sentence that a simple explanation would be simplistic if it failed to capture all the essential and relevant parts. It is essential to understand that language and meaning were still developing in the 14th Century and care in translation in relation to subject matter statements makes Ockham's statement even more incredible, for his comments crossed boundaries unlimited to specific subject matter. Ockham probably drew inspiration from earlier philosophers such as Aristotle (384–322 BC), Alhazen (965-1039), Maimonides (1138-1204), Thomas Aquinas (c. 1225–1274) and John Duns Scotus (1265–1308), the latter who Ockham, it has been suggested, studied under him at Oxford.

Essential to Ockham's philosophy, who later, it is suggested, influenced philosophers such as Francis Bacon, is an interpretation given to his work that, when arriving at a conclusion, it is on the basis that 'facts' have already been considered before a conclusion is drawn. It is that philosophy that relates to CSA practitioners, for invariably it is not the conclusion the expert arrives at but what is required to be known is how s/he got to a particular conclusion in the first place that will be tested. By way of illustration, my report in a recent case asked the question that prior to testing what enquiries were made to the mobile network operator as to what alterations had occurred at the Masts prior to conducting radio test measurements, and can the defence please have copies of the operator's written responses?

The question went to the heart of the matter regarding accuracy of test results that underpin the opinion. Significantly, it is the prosecution that deserves praise for their benchmark standard they set in the Soham Murder case of Jessica and Holly and the subsequent conviction of Mr Huntley. I was not in that case, but as I understand it for the prosecution to show how Mr Huntley had used his mobile phone required the resurrection (I believe) of a decommissioned Mast and all the other Masts with a coverage footprint illuminating towards Mr Huntley's property to have been aligned so as to be the same as it was at the material time of mobile calls. This was required, as I understand it, in order to show the mobile telephone evidence had 'weight' and 'substance' and to avoid it being kicked out of evidence were it the case that Masts were generating coverage that would be incompatible with cell coverage at the material time. I have to say I am pro-prosecution on that landmark work and it is important to give praise where it is due. It does also mean, though, that the prosecution has established a precedent for standard of workmanship for a murder case, albeit in a high profile case, and set a marker that they will work to, and would not retract from, that standard for murder cases, at least. So it is clear why I would naturally request in a murder case what checks had been made to the mobile network operator regarding changes they had made to their Masts before the prosecution expert went on to conduct tests?

No names, no pack drill, just suffice to say the defence were told operators had 'no obligation' to keep records and, if they did keep records, were found on occasions to be inaccurate so they didn't ask, was the general thrust of the response. Really! What, no requirements under the Public Mobile Operators Licence (PMOL) to retain records about a Mast up to six months after it had been decommissioned? So how on earth could OFCOM ever check matters of interference to emergency frequencies bands from an unstable Mast if operators simply ditched their records or kept unreliable records? More importantly, what does this say about historical matters?

So does the approach in that recent murder case affect the previous prosecution benchmark approach? In my opinion, No it doesn't, and I have considerably more faith in the prosecution than that. The mobile network operator's witness provided to the court evidence of logs they regularly and continuously retain about changes to their base stations. Interestingly, on and prior to the dates of radio tests being conducted in that murder case the operator had in fact been making changes at some of the Masts targeted for their cell footprint.

Now, if I am picking that up in just one case, what is happening in other cases that have or are being rammed through the Court system to hit targets and what checks have or are being made regarding accuracy?

Of equal importance, the positive aspects coming out of cases like this means we can start to build a ‘de facto’ standard as we know the things that are required to be done.

No comments: