The state of Ohio appeals from the judgment of the Athens County Municipal Court suppressing evidence of a breath-alcohol-content test administered to Colleen McNamara. The state alleges that the trial court erred by finding that Athens Police Department (“APD”) radios were within thirty feet of the breath-alcohol-content testing instrument. We disagree because the record contains substantial evidence supporting the trial court’s findings. The state also asserts that the trial court should have denied McNamara’s motion because she did not present evidence of radio traffic during her test. We disagree because McNamara presented sufficient evidence upon which the trial court could conclude that the APD radios were used during McNamara’s test. Accordingly, we affirm the judgment of the trial court.
*709I
Athens City Police Officer Jeffery McCall arrested McNamara for allegedly operating a motor vehicle while under the influence of alcohol (“OMVI”) in violation of R.C. 4511.19(A)(1) and (3). Officer McCall transported McNamara to the Ohio State Highway Patrol Post for breath testing and parked his cruiser one or two spots away from the handicap spot in the post’s parking lot. Officer McCall then entered the building, leaving his CB radio, the patrol car radio, and his cellular telephone in the “on” position in his cruiser.
Inside the building, Officer McCall saw the sign reminding officers that hand-held radios are not permitted in the room where breath-alcohol testing is conducted. At the suppression hearing he testified that, upon reading the sign, he always turns his radio off and leaves it at the front desk. However, Officer McCall could not specifically recall where he put his radio or whether he turned it off during McNamara’s test.
McNamara’s test, administered by a senior operator on a BAC Datamaster breath-analysis machine, indicated that she had consumed alcohol in excess of the legal limit. McNamara moved to suppress the test results, alleging that the results were presumptively inaccurate because APD radios, which had never been tested against the Ohio State Highway Patrol’s BAC Datamaster for potential radio frequency interference, were in use within thirty feet of the BAC Datamas-ter during her test.
Each party has identified testimony from the suppression hearing regarding the distance between the BAC Datamaster and the police cruiser in which the state concedes that three untested radios were left in the “on” position. The testimony of Sergeant Scholl of the Ohio State Highway Patrol seems to indicate that the entire parking lot is within thirty feet of the BAC Datamaster testing device. Sergeant Van Sickle testified that he measured the thirty-foot mark to reach just to the edge of the handicap spot, and that the handicap spot is the closest spot to the building. Unfortunately, due to problems with the recording equipment used at the hearing, a large portion of the testimony is not transcriba-ble.
At the hearing, the state conceded that the APD regularly used the Ohio State Highway Patrol testing device for a period of four to five months in 1996 and that the APD radios were never tested for radio frequency interference (“RFI”) during that time. The state also introduced the radio traffic log from the APD dispatcher, which indicates that no radio traffic occurred during the time that Officer McCall was at the post. The log does not reflect whether any cellular telephone calls were transmitted during that time. Cellular telephones, like *710dispatch and hand-held radios, operate on radio waves, but are not recorded in the APD radio traffic log.
The court granted McNamara’s motion to suppress, finding that APD radios had not been RFI surveyed and were within thirty feet of the BAC Datamaster testing device during McNamara’s breath test. The state now appeals, asserting that the court should not have granted McNamara’s motion. Specifically, the state frames its statement of the issue as follows:
“Does the state substantially comply with Ohio Department of Health Regulations when an Athens City Police Officer’s mobile and cellular radios left in the ‘on’ position, which had not been RFI tested, were not within 30 feet of the breath testing instrument at the Ohio State Highway Patrol Post; the officer’s hand held radio, which had not been RFI tested, was turned off and left at the front desk of the post; and there was no radio traffic reported from the time the officer arrived at the post until the time the officer left the post.”
In essence, the state asserts two assignments of error. First, the state challenges the trial court’s factual finding that APD radios were within thirty feet of the BAC Datamaster during McNamara’s test. Second, the state asserts that regardless of the location of the radios relative to the BAC Datamaster, McNamara was not prejudiced, because no radio traffic which could have interfered with the testing device occurred during her test.
II
Appellate review of a decision on a motion to suppress evidence presents mixed questions of law and fact. United States v. Martinez (C.A.11, 1992), 949 F.2d 1117, 1119. At a suppression hearing, the trial court assumes the role of trier of fact, and as such, is in the best position to resolve questions of fact and evaluate witness credibility. State v. Carter (1995), 72 Ohio St.3d 545, 552, 651 N.E.2d 965, 973. A reviewing court must accept a trial court’s factual findings if they are supported by competent, credible evidence. State v. Guysinger (1993), 86 Ohio App.3d 592, 594, 621 N.E.2d 726, 727. The trial court then applies its factual findings to the law regarding suppression of evidence. An appellate court reviews the trial court’s application of the law de novo. State v. Anderson (1995), 100 Ohio App.3d 688, 691, 654 N.E.2d 1034, 1036.
In this case, the parties dispute whether an APD radio was in use within thirty feet of the BAC Datamaster. Officer McCall has no specific recollection of turning off his hand-held radio and leaving it at the front desk at the post. However, Officer .McCall testified that it is his “normal procedure” to do so.
Additionally, each party has pointed to portions of the transcript which appear to support its own position with regard to the distance from the BAC Datamaster *711to the parking lot where Officer McCall’s cruiser held three radios in the “on” position which had not been RFI surveyed. Unfortunately, much of the testimony regarding the location of the police cruiser in relation to the BAC Datamaster is not transcribable because the trial court’s recording equipment malfunctioned. We do not know what that testimony was, but we know that upon hearing it, the trial court concluded that APD radios were within thirty feet of the BAC Datamaster testing device.
We find competent, credible evidence in the record to support the trial court’s finding that APD radios which had not been RFI surveyed were left “on” within thirty feet of the BAC Datamaster. This evidence includes both Officer McCall’s uncertainty regarding his hand-held radio and the conflicting testimony regarding the distance from the BAC Datamaster to his cruiser. Furthermore, because the record is incomplete, we must presume the trial court had adequate evidence before it to support its findings. See Knapp v. Edwards Laboratories (1980), 61 Ohio St.2d 197, 199, 15 O.O.3d 218, 219-220, 400 N.E.2d 384, 385-386; In re Adoption of Foster (1985), 22 Ohio App.3d 129, 131, 22 OBR 331, 332-333, 489 N.E.2d 1070, 1072-1073. We note that, if the state wanted us to consider the parts of the record which are incomplete, it should have followed App.R. 9(C), (D), and/or (E). Accordingly, we overrule the state’s challenge to the trial court’s factual finding that APD radios were within thirty feet of the BAC Datamaster.
Ill
We now turn to the trial court’s application of the law to McNamara’s case.' The state argues that the trial court should not have suppressed the results of McNamara’s BAC Datamaster test because the evidence presented indicated that no APD radio traffic occurred during the test. McNamara asserts that the presence or absence of radio traffic was not proven as to the cellular telephone and that radios which are “on” are “in use” as prohibited by regulation. Furthermore, McNamara asserts that the state did not trigger her burden to prove that the APD radios were used during her test, because it did not meet its initial burden of proving that it complied with regulations governing RFI testing.
The results of a breath test administered pursuant to R.C. 4511.19 may be admitted into evidence only upon a showing that the breath test was administered in accordance with Department of Health (“DOH”) regulations. Cincinnati v. Sand (1975), 43 Ohio St.2d 79, 72 O.O.2d 44, 330 N.E.2d 908, paragraph two of the syllabus. Pursuant to DOH regulations, an RFI survey must be administered with respect to each radio frequency regularly broadcast from within the vicinity of the agency’s BAC Datamaster or other approved breath-alcohol-analysis device. State v. Adams (1992), 73 Ohio App.3d 735, 743, 598 N.E.2d 176, 181-182. The RFI survey examines the existence and degree of interference created by *712radios operating near the testing device. Id. The relevant DOH regulation, Ohio Adm.Code 3701-53-02(0, provides:
“(1) Radio transmitting antennae shall not be used within any RFI affected zone during the conduct of a subject test or a calibration check. No radio transmitting antennae that have not been subjected to an RFI survey shall be used within thirty feet of the breath testing instrument during conduct of a subject test or a calibration check.”
The state need not prove strict, or perfect, compliance with DOH regulations. See State v. Plummer (1986), 22 Ohio St.3d 292, 22 OBR 461, 490 N.E.2d 902, syllabus; State v. Harris (June 29, 1995), Athens App. No. 94CA1635, unreported, 1995 WL 405004. Only “substantial compliance” is necessary for the test results to be admissible into evidence. Defiance v. Kretz (1991), 60 Ohio St.3d 1, 3, 573 N.E.2d 32, 34; Plummer at syllabus.
We must determine whether, in light of the BAC Datamaster being frequently used by the APD, whose radios had not been RFI surveyed, the state achieved substantial compliance with DOH regulations. The Champaign County Court of Appeals faced a similar question in Adams, supra, 73 Ohio App.3d 735, 598 N.E.2d 176. The Adams court declined to suppress the OMVI suspect’s breath test even though the local police department did not conduct an RFI survey on radios frequently used by the Ohio State Highway Patrol in the vicinity of the testing instrument. The court held that, absent evidence that the Highway Patrol frequencies were, in fact, transmitted from a radio within thirty feet of the local department’s testing instrument during the testing, failure to perform an RFI survey for those other frequencies did not preclude admission of appellant’s test results. Id. at 745, 598 N.E.2d at 182-183. The court explained its reasoning in two steps:
“First, because RFI can occur only if a radio frequency source is actually operated within the environment of a testing instrument, it would logically follow that one seeking to exclude evidence of breath-test results has the burden to show that a radio broadcast took place within the environment or RFI-affected zone during the time interval of the test. * * * However, the General Assembly has in R.C. 4511.19(A)(3) provided that bodily substances taken to prove a violation of the statute ‘shall be analyzed in accordance with methods approved by the director of health,’ and the director has provided in Ohio Adm.Code 3701-53-02(C) that an RFI survey conducted with the protocol prescribed ‘shall be performed’ for each approved breath-testing instrument. Furthermore, R.C. 4511.19 is a strict liability statute in which the accuracy of breath-test results are a crucial issue in determining a defendant’s guilt or innocence. Defiance v. Kretz (1991), 60 Ohio St.3d 1, 573 N.E.2d 32. Therefore, proper conduct of an RFI survey and report of its results, both originally and upon the happening of any of *713the events provided in Ohio Adm.Code 3701-53-02(0(2), is a necessary predicate to the introduction in evidence of the results of a breath test using one of the instruments approved for that purpose by the Director of Health. The state has the burden of proof concerning the issue. Cincinnati v. Sand, supra.
“Second, during the time interval of an evidential breath test * * * no radio may be broadcast at any unsurveyed frequency within the thirty-foot radius constituting the environment of the instrument. Conformity with these requirements is not, however, a foundational predicate to the introduction of breath-test results. They are not a qualifying factor. Rather, as they may affect reliability only if they occur, the burden is on the accused to come forward with evidence that either or both did occur at the relevant time, i.e., during the time interval of the accused’s test.” (Emphasis added.) Adams at 743-744, 598 N.E.2d at 182.
The state argues that it substantially complied with Ohio Adm.Code 3701- ' 53-02(C) by administering RFI surveys for the radio frequencies used by the Ohio State Highway Patrol. In support of its position, the state relies on State v. Day (1990), 67 Ohio App.3d 325, 586 N.E.2d 1190, wherein the court held that “the applicable regulations do not clearly mandate that a[n] RFI survey be conducted with respect to every frequency used by anyone who might happen to enter the premises where a machine is kept.” Day at 328, 586 N.E.2d at 1191. This court applied and expanded the Adams and Day holdings in State v. Buchanan (Nov. 22, 1996), Washington App. No. 95CA48, unreported, 1996 WL 718265. Buchanan, like Adams and Day, involved a similar fact pattern to the case before us. In Buchanan, however, the BAC Datamaster located at the local police department had not been RFI tested for radios used by the Ohio State Highway Patrol, and the Highway Patrol accounted for nearly fifty percent of the device’s use. Nonetheless, this court determined that the burden lay with the OMVI suspect to show that a radio transmitting on an untested frequency was in use within thirty feet of the BAC Datamaster when his breath test was taken. Id.
McNamara asserts that the above cases are distinguishable from the case before us. In each of the above cases, the evidence showed that no untested radio was within thirty feet of the breath-alcohol-analysis device. Additionally, in the above cases no untested radios were left in the “on” position in the vicinity of the testing device. In the instant case, the court found that untested APD radios were left in the “on” position within thirty feet of the BAC Datamaster. Based on these distinguishing factors, McNamara urges us to find that the state failed to demonstrate substantial compliance with DOH regulations. In the alternative, McNamara asserts that radios which are “on” are “in use,” and therefore she has met her secondary burden of proving that an untested radio transmitting antenna was used within thirty feet of the breath-testing instrument during her test.
*714We find that the state substantially complied with the DOH regulations regarding RFI testing in this case. For the purpose of determining substantial compliance, we find Adams, Day, and Buchanan indistinguishable from the case at hand. The fact that untested APD radios were “on” and within thirty feet of the BAC Datamaster goes to the issue of prejudice resulting from possibly incorrect test results rather than to the issue of substantial compliance.
This brings us to the question of whether McNamara met her secondary burden of proving that APD radios were actually used during her breath test. Ohio Adm.Code 3701-53-02(0) states that “[n]o radio transmitting antennae that have not been subjected to an RFI survey shall be used within thirty feet of the breath testing instrument during conduct of a subject test or a calibration check.” (Emphasis added). The regulation does not define “use.” The plain meaning of “use” is “to employ for some purpose; put into service.” Webster’s College Dictionary (Random House 1995) 1468. Prior cases have employed the terms “use,” “broadcast,” and “transmit” interchangeably without discussion. See Adams, Day, and Buchanan, swpra. As McNamara points out, however, in each of these cases the OMVI suspect failed to show that any unsurveyed radios were “on” within the vicinity of the breath-testing instrument. Whether a radio which is merely “on” is “in use” within the meaning of Ohio Adm.Code 3701-53-02(C) is an issue of first impression.
Interpretation of the regulation according to its plain meaning leads to the conclusion that a radio which is “on” is “in use,” for it is being employed for the purpose of picking up radio traffic. The reasoning employed in Adams and adopted by this court in Buchanan provides a more persuasive, but not necessarily inconsistent, method to determine the meaning of “use.” Pursuant to Adams, an OMVI suspect suffers no prejudice from an unsurveyed radio frequency source unless the source is actually emitting radio waves during the suspect’s breath test. Adams, 73 Ohio App.3d at 743, 598 N.E.2d at 181-182. Whether radio waves are emitted from a radio which is turned “on” but registers no radio traffic is a question of fact, best resolved by the trier of fact. Thus, the trier of fact, probably with the help of expert testimony, must determine whether a radio which is “on” is “in use.”
The trial court did not make a finding as to whether the APD radios emitted radio waves or were “used” during McNamara’s test. Where the trial court has failed to issue a finding of fact, we look directly to the record to determine “if there is sufficient evidence demonstrating that the trial court’s decision was legally justified and supported by the record.” State v. Brown (1992), 64 Ohio St.3d 476, 597 N.E.2d 97, syllabus. The only evidence in the record supporting a finding that the APD radios were in use during McNamara’s test was the testimony that the radios were left in the “on” position. However, *715as noted above, much of the record is not available to us due to an equipment malfunction at the suppression hearing. Because the record is incomplete, we must presume the trial court had adequate evidence before it to support its findings. See Knapp, supra, 61 Ohio St.2d at 199, 15 O.O.3d at 219-220, 400 N.E.2d at 385-386.
This court will not substitute its own judgment for that of the trial court. Berk v. Matthews (1990), 53 Ohio St.3d 161, 169, 559 N.E.2d 1301, 1308-1309. Based on the record before us, we cannot say that radios which are “on” are not “in use.” We find that the plain meaning of the regulation dictates that the state violates the regulation when unsurveyed radios are left “on” in the vicinity of the breath-testing instrument. We further find that McNamara presented sufficient evidence upon which the trier of fact could reasonably conclude that untested APD radios were left “on” within the vicinity of the BAC Datamaster during her test. We also presume, based on the trial court’s decision and an incomplete record, that McNamara made a sufficient showing that the unsurveyed APD radios emitted radio waves within a thirty-foot radius of the BAC Datamaster during her breath test. Therefore, the results of her breath test may not be introduced into evidence. Adams, supra, 73 Ohio App.3d at 744, 598 N.E.2d at 182.
IV
In conclusion, we find that the state met its burden to show that it complied with DOH regulations requiring the proper conduct of an RFI survey with respect to the Ohio State Highway Patrol BAC Datamaster located at the patrol post on Columbus Road in Athens County, Ohio. Further, we find that McNamara made a sufficient showing of prejudice by showing that unsurveyed APD radios were used within a thirty-foot radius of the BAC Datamaster during her breath test. Therefore, the trial court did not err in suppressing the breath-test results. Accordingly, we overrule the state’s assignment of error.
Judgment affirmed.
Stephenson, P. J., concurs.
Haesha, J., concurs in judgment only.