The invention relates to the field of seismic and/or radar remote sensing, in particular for the location of deep underground objects and/or structures. More specifically, the invention relates to a method of identifying reflected signals during such remote sensing operations.

Exploration for earth resources using pulsed waves are widely used in industry. Such methods are limited by the resolution of the reflections which is determined by distance and clutter and noise. A widely used method (called "stacking") is to repeat the measurement several times and to average the resulting measured return signals. In such a case the reflections will add up in the averaging, but the noise will not, resulting in higher resolution.

It is desirable to provide a method to identify reflections which are so weak that they are not apparent even after stacking thus extending the effective range of the pulsed wave method of exploration. SUMMARY OF INVENTION

In a first aspect of the invention there is provided a method of identifying reflected signals, subsequent to their reflection within a medium, said method comprising: obtaining a plurality of return signals, resulting from a plurality of measurements being performed over a measurement period, said measurement period comprising plural sub-periods, said return signals comprising reflected signals and noise;

partitioning the plurality of return signals into plural sets of equal cardinality or as equal as possible such that their cardinality differs by no more than one;

determining a stacked correlation value for said return signals by determining the mean of said return signals across said plural sets and determining a correlation value of the plural sets over each of said time sub-periods; and identifying peaks in the variation of said stacked correlation value over time and attributing each of said peaks in the variation of said stacked correlation value over time to a reflected signal. Other aspects of the invention comprise a computer program comprising computer readable instructions which, when run on suitable computer apparatus, cause the computer apparatus to perform the method of the first aspect; and an apparatus specifically adapted to carry out all the steps of any of the method of the first aspect. Other non-essential features of the invention are as claimed in the appended dependent claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described, by way of example only, by reference to the accompanying drawings, in which:

Figure 1 is a flowchart describing a method according to an embodiment of the invention;

Figure 2 is a flowchart describing in more detail a calculation step of the method of Figure 1; and

Figure 3 is a graph showing an exemplary result of the method of Figure 1.

DETAILED DESCRIPTION OF THE EMBODIMENTS

Disclosed is a method to identify weak reflections of a wave pulse sent into a region for identifying objects and/or features. The wave pulse can be an acoustic pulse or an electromagnetic pulse. In a typical application an electromagnetic or acoustic pulse is sent into the ground from the surface and the return signal is measured at the surface. The return signal is then analysed for the presence of reflections from structures below ground, which can then be located if the (average) wave propagation velocity is known.

In some cases, the return signals are sufficiently weak that the reflections are obscured by clutter and/or noise and are not identifiable by conventional means. This typically happens when trying to explore deep regions in the earth. Some reflections may be so weak that they are not apparent even after stacking. Such weak reflections occur from far objects and/or structures. The ability to identify them will result in a significant increase in exploration depth using the pulsed wave method. Figure 1 is a flowchart describing a method according to an embodiment of the invention. The method applies to circumstances where wave pulses are repeatedly (N times, where JV may take any value; for example, JV may lie between 10 and 5000, or between 100 and 1000; JV = 500 is a typical number) sent into the medium being explored. At step 100, the return signals are measured for each pulse. These return signals and are denoted by Xi{t], where i = 1, . . . ,N labels the specific pulse, 0 < t < T represents time, and T is the duration of the measurement. In most applications t is discretely sampled. The aim is to identify reflections which are present in all JV signals, but are too faint to be directly visible by a visual inspection of a graph of Xi(t).

At step 110, the time interval [0 T] is partitioned into K sub-periods. This partition may be determined by the time sequence to < ti < . . . < ίκ-ι < t _¾ with to = 0, ίκ = T, and the other times can be chosen as appropriate. The kth sub-period is denoted by = [tk-i tk] where τ = ½(ί¼-ι +tk). The choice of partition is application dependent and is typically chosen to have sub-periods that are at least as large as the temporal extent of the expected reflection signal.

Following this, and for each sub-period, three parameters are computed.

At step 120, the first of these parameters is computed, which is herein called the "raw correlation". This is the mean over all distinct signal pairs x,(t), Xj(t), i≠ j of their correlation over the sub-period.

In general, where a correlation is calculated as part of a method disclosed herein, the correlation C[y, z, k) may be calculated as follows; where y{t) and z{t) are the two signals being correlated over a sub-period Δ^:

where

It should be understood that one or more of the integrals may be replaced by a discrete approximation in standard fashion should time t be sampled discretely (as will usually be the case).

Specifically, the "raw correlation" % may be calculated at step 120 for each time interval k [l≤k≤K) as follows:

¾ =χ , , ί)/^-ι) At step 130, the second of these parameters, called herein the "stacked correlation" is calculated. Figure 2 is a flowchart expanding on the method of this step.

At step 220, the set of N signals is randomly divided into two disjoint sets of equal size N/2 if N is even, or of sizes (N +l)/2 and (N -l)/2 if N is odd. At step 230, the mean signals across both sets are computed. At step 240 their correlation across each time sub-period is computed. This procedure is repeated M times (step 250) after which the results are averaged (step 260).

More specifically, step 130 may comprise letting [//] denote a random partition of 1, . . . ,N into two disjoint sets / and / with equal size or cardinality denoted by |/| and |/| if N is even (i.e., |/| = |/|) and with / containing one more element than / if N is odd (i.e., I/I = |/| +1). M distinct random partitions may be generated, denoted by [I _m Jm], m = 1, . . . ,M. M may be any value, for example M may lie between 10 and 1000, or 10 and 500. A typical value of M is 100. "Stacked correlation" Sk is defined for each time interval k [l≤k≤K) as follows:

M

S _k =∑c _mk /M

m-1

where

At step 140, the third parameter, called herein the "stacked standard deviation" is calculated. This comprises calculating, for each time sub-period, the standard deviation of the quantity that was averaged as described in the previous paragraph. Consequently, the "stacked standard deviation" Dk for each time interval k (l≤k≤K)is defined as follows: At step 150, the three parameters Rk, Sk, and Dk are plotted versus the centre time Tk of the time sub-periods. At step 160, reflections are identified by 1) peaks in the raw correlation, or 2) peaks in the stacked correlation which lie above the stacked standard deviation. It is the latter that has the ability to detect very faint reflections; the higher they lay above the stacked standard deviation the more significant they are.

Figure 3 is a graph showing an exemplary result, based on simulated data. The horizontal axis represents time from 0 to about β. ΒμΞ. Plot 310 is a plot of the stacked correlation, plot 320 is a plot of the raw correlation, and plot 330 is the stacked standard deviation. A peak 350 can be seen on the raw correlation plot 320 at around t = Ι,βμΞ, representing a strong reflection. Two peaks 340, 360, each above stacked standard deviation plot 330, can be seen on the stacked correlation plot 310 at around t = 3.5 5 and t = 6.2 These peaks 340, 360 indicate weak reflections. One or more steps of the methods and concepts described herein may be embodied in the form of computer readable instructions for running on suitable computer apparatus, or in the form of a computer system comprising at least a storage means for storing program instructions embodying the concepts described herein and a processing unit for performing the instructions. As is conventional, the storage means may comprise a computer memory (of any sort), and/or disk drive, optical drive or similar. Such a computer system may also comprise a display unit and one or more input/output devices.

The concepts described herein find utility in all aspects (real time or otherwise) of identification, surveillance, monitoring, optimisation and prediction of a subsurface volume. It may be used, purely by way of example, in the identification, surveillance, monitoring, optimisation and prediction of a hydrocarbon reservoir, and may aid in, and form part of, methods for extracting hydrocarbons from such hydrocarbon reservoir and well systems.

It will thus be seen that the objects set forth above, among those made apparent from the preceding description, are efficiently attained and, because certain changes may be made in carrying out the above method and in the construction(s) set forth without departing from the spirit and scope of the invention, it is intended that all matter contained in the above description and shown in the accompanying figure shall be interpreted as illustrative and not in a limiting sense.